Machine Learning Theory, AI Safety/Alignment, Optimization
Kaifeng Lyu is an Assistant Professor at the Institute for Interdisciplinary Information Sciences (IIIS), Tsinghua University. His research primarily focuses on theoretical and safety problems in deep learning and large language models. Before joining Tsinghua University, he was a postdoctoral research fellow at the Simons Institute for the Theory of Computing at UC Berkeley. He received his Ph.D. in Computer Science from Princeton University in 2024, where he was advised by Prof. Sanjeev Arora, and his B.Eng. from Tsinghua University in 2019. His work has appeared at leading machine learning conferences including NeurIPS, ICML, and ICLR, and was recognized with an ICLR 2025 Best Paper Award.
StudentsPhD Students:
Haodong Wen (2025–present)
Kexian Tang (2025–present)
Huaijie Wang (2023–present, joined our group in 2026)
Jinhan Li (2026–present)
Tingkai Yan (2026–present)
Master's Student:
Rui Chen (2025–present)
TeachingCourses
Spring 2026. Mathematics for Computer Science and Artificial Intelligence, Tsinghua University.
Fall 2025. Large Language Models from Scratch: Theory and Practice, Tsinghua University.
Related information